After 2020, there are only two types of businesses: those who have retained customers or grown, and those who lost customers to more nimble competitors. There are many reasons. The transition away from traditional retail, shifting customer expectations, and internal operations challenges due to increasing complexity all play their parts. However, the main reason is simple. 2020 affirmed that there’s no such thing as a non-digital business.
Even minor gaps between digital experience expectations and actual service delivery can be visible to all. Users, employees, partners, and innumerable service integration interfaces must all perform to the very limit of available budget. Google makes competitors instantly discoverable, and immediate side-by-side comparison has become the norm.
How many mobile apps have you replaced on your phone simply because another was faster, more smoothly integrated, or simply more enjoyable
to use? Customers now expect they can use mobile devices to get everything they can normally get from your office, website, or other channels. But can they? If the competition’s digital experiences are better and further improving, their technology team likely realized this. You’re watching their ROI in action.
It can seem as if suddenly almost everyone—IT pros, architects, managers, CIOs, and more—are making previously unimaginable success statements. There’s legitimate glee or at least relief: “We’d planned a six-month O365 migration but finished in two weeks!” “Touchless? We just extended service for POS dongles, done.” “Component sourcing dried up, so we doubled our suppliers and got production lines back up to speed.” “Standing up the new SaaS app only took two admins a few days. Who knew?”
If those sound familiar, you may be part of an organization focused on digital experience since 2019 or earlier. When the world turned upside down, and work, school, family, and even recreation demanded timely transformation, your team may have already made the leap. You made assuring great digital experiences an equal operations citizen on par with optimizing networks and applications. Central to this shift, successful technology teams extended traditional infrastructure monitoring to add digital experience monitoring to their toolbelts. That combination delivered end-to-end performance visibility tuned for modern applications.
Feelz-Driven Root Cause Analysis
One of the challenges of end user experience is technologists don’t like feelings at work. Clarification: operations—especially at enterprise scale—can be overflowing with feelings. As a result, we seek hard data wherever possible to escape the multivariant conjectures of emotions-led, reactive decision-making. Hunches sometimes
lead to IT eureka, but more often they distract from logical, repeatable
paths to resolution. Major infrastructure or application design changes based on user’s feelings is a departure from years of ops processes. It’s natural to be skeptical. The trick is to do what IT pros do best. Convert opinions into quantifiable data, and users into ultimate performance monitoring agents.
Consider a practical example. Employees in a West Coast field office are complaining an application is crawling. Of course, they’re not complaining via service desk
tickets, only to each other. The application is “useable,” but in an ugh, not again, might as well be down, groan-inducing way. (I still recommend setting up a context search-bot for Teams/Slack to alert on common phases your employees use when CSAT heads downhill, but that’s for another article.) Without digital experience visibility, the resulting troubleshooting misadventure is likely familiar:
- The initial service desk ticket comes from an internal IT director who was forwarded an email from her peer in Sales, hours or days after the issue started.
- Tier 1 support breaks out the infrastructure monitoring dashboard and looks at histograms, alerts, and logs for performance issues. Nothing jumps out for the app’s components, resources, or network. Perhaps it’s the internet maybe? They give it time.
- The ticket is escalated to Tier 2 when Sales escalates the email thread to an IT VP.
- Senior admins verify the back end looks fine but add more compute and network headroom anyway, just to be sure.
- The issue escalated to Tier everybody when the head of Sales mentions the “long-running issue” to the CEO at lunch, who mentions it to the CIO. Then it’s all hands on deck, productivity hits pause, tensions rise, the clock runs.
- “New kid” randomly mentions a firewall is reporting VPN tunnel issues. Heads slowly swivel to them. “What?” asks the pacing IT director.
- Your most senior admin replaces an expired certificate on a West Region security appliance. The Fresno Sales director sends a Slack message to ask what you did and says the issue is resolved.
- You close the ticket, now with pages of troubleshooting details from the dozen people involved.
How About a Do-Over?
Service issues like that are rare, challenging to define alerts for, and build fire alarm panic backpressure that can sideline everyone without notice. Yet most IT pros have been in those situations more times than they care to remember. Worse, those types of incidents comprise the bulk of the reactive distractions for most teams. Instead, consider a different response to the same incident when digital experience monitoring is available.
- Half the branch office VPN bandwidth supporting a hybrid application goes offline, but no network devices go into a red, down status. A small subset of app server transaction duration statistics spike but are muxed with overall infrastructure performance, along with other transient issues. The NMS system does catch a certificate expiration error from a VPN. Aka, this issue is super edge case, novel, and unlikely to pop out on a NOC view.
- However, the team previously added user experience monitoring for the application and is collecting metrics that suddenly 15% of users are seeing page load times well above baseline.
- The Sales team in a regional office decides to catch up on fantasy football, no tickets or emails are sent.
- A user experience alert fires and a red dot appears as a sub-status warning for the application on the NOC view.
- “New kid” pulls up the composite view of the application in the monitoring dashboard, notes the average latency is up 10%. They also notice a correlated certificate expiration alert to the left of the sustained rise in the chart. “Aha,” they say and group the chart by app region. There it is. Everything is normal except for one office where everyone is having a bad day. They open a ticket for the network security team.
- Ten minutes later, the certificate is updated, the app goes green again on the dashboard, and Sales is back to closing deals.
In this example, it’s a feelings-driven issue. There was no outage in the classic sense, but one sales team in one office felt frustrated by lack of response to an issue unique to them. Hearing, “Well, we don’t see it on our end,” only adds to it. Sales management feels they’re not getting the support they need, and the CIO feels concern there’s a larger application delivery challenge of unknown scope. Feelings, feelings, feelings. Digital experience monitoring can keep operations and even developers well ahead of that. It proactively converts future upset into latency, error, usage, engagement and other metrics actionable at the time of the problem. Experience monitoring is human behavioral precognition for proactive incident mitigation.
Users. You Keep Using That Word.
There’s one last but critical strategy for closing the gap between your user’s expectations and the digital experiences they receive. It’s less intuitive but may also be the one a competitor addressed first. Stop thinking of users as people. Customers, employees, and partners are most certainly humans who will be the ultimate judges at the end of transaction pipelines. You’re striving to ensure people are delighted. However, with modern applications, component interconnect issues can have a greater impact on service quality than any other factor.
Some teams make this discovery almost by accident. They stop categorizing and prioritizing “users” by type. Human, customer, partners, APIs, etc., are all treated as service consumers
in a mind-shift to transactional equality. Categorization once made sense because IT pros had complete control of all back-end resources and (most) delivery paths. Categories aligned with infrastructure topology and application elements. It was handy for debugging.
Today however, developers are rushing to reimplement and deconstruct vertically aligned app stacks into distributed services. They’re intentionally divorcing tailored application component connections in the search for smaller services that are easier to evolve and reuse. The result is an increasingly matrixed web of other services whose demand diversity rivals the user groups who depend on them. Modern apps stack their data calls high and wide, adding multiple transactional boundary latencies unknown to their monolithic forebears.
As distribution, interconnection, and reuse rise, troubleshooting efficiency decreases if performance monitoring is infrastructure only. The final consumers of applications and the resources they depend on remain—as always—the ultimate judges of operations quality. It’s their perceptions which measure the business as a whole.
Successful competition today presents an interesting paradox. By taking the stereotype of “users” out of the equation, it’s easier for IT pros to drive loyalty for a business. Humans will reward you when they feel you care about their digital experiences, wherever in your operations that may lead.